krona

22 ms·

301.
by krona 3y ago
Causal profiling still uses hypothesis testing, so I don't get your point. Ultimately, one needs to know if software A is 'better' or 'worse' than software A' for any given definition of those terms.
302.
Geoffrey Hinton – Two Paths to Intelligence(youtube.com)
2 points by krona 3y ago|1 comments
303.
by krona 3y ago
Or, compile with -fsanitize=undefined...
304.
by krona 3y ago
In the UK, 21% of all cancer deaths are from lung cancer. Surely this is the more appropriate way to assess efficacy in the context of all cancers.
305.
by krona 3y ago
30 people seems to be far too small of a sample size for any meaningful results, except to conclude the usual "a larger study is needed to confirm or disprove our findings" funding gambit.
306.
by krona 3y ago
Because it decreases social trust, the bedrock of a functioning society.
307.
by krona 3y ago
Isn't this Office for National Statistics data?
308.
by krona 3y ago
I was just about to say the same thing. Unfortunately I think it's so deeply ingrained in Swedish culture (Law of Jante etc.) that it's hard for most Swedes to even realise there's a problem, or at least that institutional co
309.
by krona 3y ago
If we're talking about mortality salience, then there's quite a lot of evidence that mortality salience predicts a greater desire to have children, which is the opposite to what you're suggesting. From an evolutionary point o
310.
by krona 3y ago
The Revolut UK banking chief executive also just quit 'for unrelated reasons' after texting a death threat to a customer: https://www.telegraph.co.uk/business/2023/05/13/revolut-exec...
311.
by krona 3y ago
> Is there generally applicable, widely accepted principles or research around this J. Ousterhout is gaining traction, at least in my corner of the industry. https://web.stanford.edu/~ouster/cgi-bin/cs190-winter
312.
by krona 3y ago
I've heard of the Grenfell Tower Fire, that was a great fire in London.
313.
by krona 3y ago
> because people before you were dumb as fuck There are many Tudor farm houses built out of dried cow/horse shit and hair where this hoard was found, and school children visit these 500 year old buildings daily to experience the eng
314.
by krona 3y ago
The trend is a form of preferential attachment. https://en.wikipedia.org/wiki/Preferential_attachment
315.
by krona 3y ago
Isn't the problem that if the risk is entirely hypothetical, you can't define a limiting principal to any hypothetical regulation? It reminds me of the Vulnerable World Hypothesis - that at any moment someone could invent somethin
316.
by krona 3y ago
Isn't the water table too high at that location? If that's true then it was never a realistic possibility to build a trench.
317.
by krona 3y ago
Running within a linux cset shield is a fairly standard practice. For benchmarks reporting times in the range of nanoseconds a common approach is a linear regression of varying batch sizes; I'm not sure gtest does this. But generally,
318.
by krona 3y ago
> There are a lot of export commodities they no longer have a market for Exports to the EU hit a record high in the summer of 2022, which ironically, was mainly because of commodities. Who'd have thought? https://www.ft.c
319.
by krona 3y ago
LOL citation needed. (Yes, everyone knows about the wife.)
320.
by krona 3y ago
Metamorphic testing seems to try to map an output of a model to a ground truth, which I guess is great if you have a database of all the known truths in the universe.
321.
by krona 4y ago
> It can only imitate. That already eclipses (at a guess) 75% of humanity, then.
322.
by krona 4y ago
Software is driven entirely by utility value, not production value. I don't care whether the app was written on stone tablets in hand-rolled assembly, or by an AI, so long as the plane lands safely.
323.
by krona 4y ago
I think a 'charitable' take would be that you don't understand what I'm saying, at all. That's my charitable take.
324.
by krona 4y ago
I don't have the US statistics on hand, it was merely a reference to the fact that 1. Most people seem to think anti-semitism was invented in the 1930's, and 2. anti-semitism defines your politics. Neither are true.
325.
by krona 4y ago
There's nothing requiring non-determinism in these models, it's just there as a means to increase variety. There are obviously scenarios where this makes less sense.
326.
by krona 4y ago
Their behaviour could be grouped in ways not derived primarily by sex?
327.
by krona 4y ago
Meanwhile, Goldman strategists are predicting this quarters earning season will be the worst for profits/EPS since the start of the pandemic. > Earnings per share are expected to decline 7% year-over-year, according to a note from
328.
by krona 4y ago
Police deter crime by increasing the perception that criminals will be caught and punished. The evidence is so overwhelming I don't know where to begin. How about https://nij.ojp.gov/topics/articles/five-thing
329.
by krona 4y ago
> Humans are lazy I don't think you realize how easy it is if you know how to track and jog for 2-3 hours for antelope, even less for cats. All you need is a club. 'Running' isn't necessary.
330.
by krona 4y ago
> Creativity should not be part of infrastructure efforts on a level beyond embellishments and experiments. You could say the same about architecture, but only modern architects think that way, because they were brainwashed by fashion.

More ›